Processing control method and device for injection product, computer equipment and storage medium

Abstract

The embodiment of the invention relates to the technical field of injection molding control, and discloses a processing control method of an injection molding product, which comprises the following steps: acquiring current operating parameters of injection molding equipment and a surface image of a current injection molding product shot at an outlet of the injection molding equipment, and inputting the surface image into a bubble recognition model to obtain a bubble recognition result; when the bubble recognition result is that bubbles exist, acquiring bubble image features in the surface image; inputting the bubble image features and the current operation parameters into a bubble cause analysis model to obtain a bubble cause and a bubble elimination strategy of the current injection product; and the machining process of the injection molding equipment is adjusted according to the bubble elimination strategy. Through the above mode, the embodiment of the invention realizes accurate identification of bubble causes in the processing process of the injection molded part, and reduces the probability of defective products.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of injection molding control technology, and specifically to a processing control method for an injection molding product, a processing control device for an injection molding product, a computer device, and a computer-readable storage medium. Background Art

[0002] At present, with the rapid development of science and technology, manufacturers pay extensive attention to the efficiency of product defect detection and processing control process in the production process.

[0003] Injection molding products are widely used in our daily life. However, bubbles may be generated on the surface of the products during the processing. There are many reasons for the generation of bubbles, such as too fast injection speed, insufficient temperature control, etc.

[0004] However, most of the existing technologies rely on manual inspection of defects in injection molded products, but it is difficult to accurately identify the cause of bubbles manually, and automatic processing control cannot be achieved. There are also methods in the existing technology that determine the cause by performing image analysis on bubbles, but the accuracy of the technical method for cause analysis is low. Summary of the invention

[0005] In view of the above problems, an embodiment of the present invention provides a processing control method for an injection molded product, a processing control device for an injection molded product, a computer device and a computer-readable storage medium, which are used to solve the problems in the prior art of inaccurate analysis of the causes of bubbles and the inability to automatically control the process of injection molded product processing.

[0006] According to one aspect of an embodiment of the present invention, a processing control method for an injection molded product is provided, the method comprising: Acquire the current operating parameters of the injection molding equipment and the surface image of the current injection molding product captured at the injection molding equipment outlet; wherein an image capturing hood is provided at the injection molding equipment outlet, and a camera device and a plurality of light sources at different angles and color temperatures are provided in the image capturing hood; the camera device is used to acquire the surface image of the current injection molding product; The surface image is input into a bubble recognition model to obtain a bubble recognition result; the bubble recognition result includes the presence or absence of bubbles; the bubble recognition model is obtained by training injection molded product bubble samples according to an image recognition algorithm, and the injection molded product bubble samples include multiple types of injection molded product bubble sample images and corresponding bubble type labels; When the bubble recognition result is that bubbles exist, acquiring bubble image features in the surface image; The bubble image features and the current operating parameters are input into the bubble cause analysis model to obtain the bubble cause and bubble elimination strategy of the current injection molded product; the bubble cause analysis model is obtained by inputting a multimodal fusion network training based on bubble parameter samples; the bubble parameter samples include bubble image feature samples and corresponding operating parameter samples, as well as corresponding bubble cause labels and bubble elimination strategy labels; The processing of the injection molding equipment is adjusted according to the bubble elimination strategy.

[0007] In an optional manner, the obtaining of the current operating parameters of the injection molding equipment and the surface image of the current injection molded product photographed at the outlet of the injection molding equipment includes: Acquire the real-time control parameters of the control system of the injection molding equipment and the real-time operation parameters of the equipment when processing the current injection molding product as the current operation parameters; the real-time control parameters include injection speed, holding pressure, cooling time and mold temperature; the real-time operation parameters of the equipment include the temperature, pressure, motor speed and hydraulic system pressure of each equipment component; When the injection molding of the current injection-molded product is completed, the surface images of the current injection-molded product at multiple angles are acquired by the photographing device.

[0008] In an optional manner, before inputting the surface image into the bubble recognition model to obtain the bubble recognition result, the method includes: Acquire injection molded product bubble samples; the injection molded product bubble samples include multiple types of injection molded product bubble sample images and corresponding bubble type labels; the bubble type labels include water bubbles, additive bubbles, poor exhaust bubbles, temperature uneven bubbles, injection speed bubbles, insufficient pressure holding bubbles, cooling time bubbles, equipment screw wear bubbles and equipment barrel temperature bubbles; Inputting the injection molded product bubble sample into the image recognition algorithm to obtain a predicted bubble type; Calculating a loss of the image recognition algorithm based on the predicted bubble type and the bubble type label; Adjusting the parameters of the image recognition algorithm according to the loss, and continuing to input the injection molded product sample into the image recognition algorithm after the adjusted parameters for training, and continuing to calculate the loss according to the new predicted bubble type and the bubble type label, until the loss is less than a preset loss threshold, thereby obtaining a trained bubble recognition model; The step of inputting the surface image into a bubble recognition model to obtain a bubble recognition result includes: Inputting the surface image into a bubble recognition model to obtain a target bubble type; The bubble recognition result is determined according to the at least one target bubble type.

[0009] In an optional manner, the image recognition algorithm is a convolutional neural network, and the convolutional neural network includes a convolutional layer, a fully connected layer, and an intermediate layer; when the bubble recognition result is that bubbles exist, obtaining bubble image features in the surface image includes: When the bubble recognition result is that bubbles exist, the output of the middle layer of the convolutional neural network is used as the bubble image feature.

[0010] In an optional manner, the bubble image features and the current operating parameters are input into a bubble formation analysis model to obtain the bubble formation cause and bubble elimination strategy of the current injection molded product, including: Splicing the bubble image features and the current operating parameters to obtain splicing feature data; The splicing feature data are fused and classified to obtain the bubble cause and bubble elimination strategy of the current injection molding product.

[0011] In an optional manner, the elimination strategy includes at least one control parameter adjustment strategy, at least one equipment operation parameter adjustment strategy and at least one material adjustment strategy. The control parameter adjustment strategy includes exhaust control strategy, temperature control, injection speed control strategy, pressure holding control strategy and cooling time strategy. The equipment operation parameter adjustment strategy includes equipment screw adjustment strategy and equipment barrel adjustment strategy. The material adjustment strategy includes moisture adjustment strategy and additive adjustment strategy.

[0012] In an optional manner, the obtaining of the current operating parameters of the injection molding equipment and the surface image of the current injection molded product photographed at the outlet of the injection molding equipment includes: Decrypting the current operating parameters according to a preset decryption method to obtain the decrypted current operating parameters; The decrypted current operating parameters are normalized to obtain normalized current operating parameters. According to another aspect of an embodiment of the present invention, a processing control device for an injection molding product is provided, comprising: A first acquisition module is used to acquire the current operating parameters of the injection molding equipment and the surface image of the current injection molding product captured at the outlet of the injection molding equipment; wherein an image capturing hood is provided at the outlet of the injection molding equipment, and a camera device and a plurality of light sources at different angles and color temperatures are provided in the image capturing hood; the camera device is used to acquire the surface image of the current injection molding product; A recognition module, used for inputting the surface image into a bubble recognition model to obtain a bubble recognition result; the bubble recognition result includes the presence or absence of bubbles; the bubble recognition model is obtained by training injection molded product bubble samples according to an image recognition algorithm, and the injection molded product bubble samples include multiple types of injection molded product bubble sample images and corresponding bubble type labels; A second acquisition module, configured to acquire bubble image features in the surface image when the bubble recognition result indicates that bubbles exist; An analysis module is used to input the bubble image features and the current operating parameters into a bubble cause analysis model to obtain the bubble cause and bubble elimination strategy of the current injection molded product; the bubble cause analysis model is obtained by inputting a multimodal fusion network training based on bubble parameter samples; the bubble parameter samples include bubble image feature samples and corresponding operating parameter samples, as well as corresponding bubble cause labels and bubble elimination strategy labels; An adjustment module is used to adjust the processing process of the injection molding device according to the bubble elimination strategy. According to another aspect of an embodiment of the present invention, a processing control device for an injection molding product is provided, comprising: A processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to execute the operation of the processing control method of the injection molding product.

[0013] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is provided, wherein at least one executable instruction is stored in the storage medium. When the executable instruction is executed on a processing control device for an injection molding product, the processing control device for the injection molding product executes the operation of the processing control method for the injection molding product.

[0014] The embodiment of the present invention obtains the current operating parameters of the injection molding equipment and the surface image of the current injection molded product taken at the outlet of the injection molding equipment, and inputs the surface image into a bubble recognition model to obtain a bubble recognition result; when the bubble recognition result is that bubbles exist, the bubble image features in the surface image are obtained; the bubble image features and the current operating parameters are input into a bubble cause analysis model to obtain the bubble cause and bubble elimination strategy of the current injection molded product; the processing process of the injection molding equipment is adjusted according to the bubble elimination strategy. In this way, various types of surface bubbles can be accurately identified, thereby tracing the causes of bubbles in the processing process, so that the processing process of the injection molded product can be controlled and the probability of defective injection molded products can be reduced.

[0015] The above description is only an overview of the technical solution of the embodiment of the present invention. In order to more clearly understand the technical means of the embodiment of the present invention, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the embodiment of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The accompanying drawings are only used to illustrate the embodiments and are not to be considered as limiting the present invention. In addition, the same reference symbols are used to represent the same components throughout the accompanying drawings. In the accompanying drawings: Figure 1 A schematic flow chart of a method for controlling the processing of an injection molded product provided by an embodiment of the present invention is shown; Figure 2 A schematic diagram of a sample image of bubbles of an injection molded product in a processing control method of an injection molded product provided by an embodiment of the present invention is shown; Figure 3 A schematic diagram showing the structure of a processing control device for an injection molded product provided by an embodiment of the present invention is shown; Figure 4 A schematic structural diagram of a processing control device for an injection molded product provided in an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0017] Exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be limited to the embodiments set forth herein.

[0018] Figure 1 The flowchart of the processing control method of the injection molding product provided by the embodiment of the present invention is shown, and the method is executed by the injection molding product processing control device. The injection molding product processing control device can be a device with control and data processing functions, such as a computer device, an injection molding product processing remote control center, a cloud processing device, etc., and the embodiment of the present invention does not make specific restrictions. Figure 1 As shown, the method comprises the following steps: Step 110: Acquire the current operating parameters of the injection molding equipment and the surface image of the current injection-molded product photographed at the outlet of the injection molding equipment.

[0019] Among them, the embodiment of the present invention obtains the real-time control parameters of the control system of the injection molding equipment and the real-time operating parameters of the equipment when processing the current injection molded product as the current operating parameters; the real-time control parameters include injection speed, holding pressure, cooling time and mold temperature; the real-time operating parameters of the equipment include the temperature, pressure, motor speed and hydraulic system pressure of each equipment component.

[0020] Since the injection molding product processing control device based on the method of the embodiment of the present invention may be a cloud processing device, in order to avoid the leakage of processing data, the control system of the injection molding device of the embodiment of the present invention first encrypts the current operating parameters according to a preset encryption method before sending them to the injection molding product processing control device. After receiving the current operating parameters, the injection molding product processing control device decrypts the current operating parameters according to a preset decryption method to obtain the decrypted current operating parameters; and normalizes the decrypted current operating parameters to obtain the normalized current operating parameters. Through normalization, the current operating parameters can be in the same dimension.

[0021] Wherein, when the injection molding of the current injection molded product is completed, the surface images of the current injection molded product at multiple angles are obtained by the shooting device. In an embodiment of the present invention, an image shooting hood is provided at the outlet of the injection molding equipment, and a camera device and multiple light sources at different angles and color temperatures are provided in the image shooting hood; the camera device is used to obtain the surface image of the current injection molded product. Specifically, since the surface of the injection molded product may have a curved surface, or due to its material, it may be illuminated and reflected by the light source, in order to avoid the generation of reflective points during shooting, the embodiment of the present invention is provided with an image shooting hood, which is covered around the camera device, and the inner wall of the image shooting hood is a weakly reflective or light-absorbing material. Wherein, in order to further avoid reflections, the embodiment of the present invention is also provided with multiple light sources, and at the same time, a filter is used to effectively filter out the reflective points. Wherein, after obtaining the surface image, the image is also pre-processed to obtain a surface image of uniform size and clarity, so as to facilitate subsequent image recognition and prediction.

[0022] Step 120: input the surface image into a bubble recognition model to obtain a bubble recognition result.

[0023] The bubble recognition result includes the absence of bubbles or the presence of bubbles. The bubble recognition model is obtained by training injection molded product bubble samples according to an image recognition algorithm, and the injection molded product bubble samples include multiple types of injection molded product bubble sample images and corresponding bubble type labels.

[0024] In the embodiment of the present invention, in order to obtain a bubble recognition model, the image recognition algorithm is trained in the following manner.

[0025] Specifically, the steps include: Step 001: Obtain injection molded product bubble samples; the injection molded product bubble samples include multiple types of injection molded product bubble sample images and corresponding bubble type labels; the bubble type labels include water bubbles, additive bubbles, poor exhaust bubbles, uneven temperature bubbles, injection speed bubbles, insufficient pressure holding bubbles, cooling time bubbles, equipment screw wear bubbles and equipment barrel temperature bubbles. Among them, since an injection molded product may exist at the same time due to various reasons such as moisture, additives, poor exhaust, uneven temperature, too fast injection speed, insufficient pressure holding, too short cooling time, equipment screw wear and too high equipment barrel temperature, bubbles at different positions in an injection molded product bubble sample image may correspond to different labels.

[0026] Among them, Figure 2 As shown, in the embodiment of the present invention, a variety of injection molded product bubble sample images are collected in advance. By collecting historical bubble images and combining them with changes in production parameters to test and generate a variety of bubble images, the injection molded product bubble samples can be established. Through expert analysis, the reasons corresponding to various bubble images are obtained, and the injection molded product bubble sample images are labeled with bubble type labels. Figure 2The upper left corner shows bubbles caused by too fast injection speed, and the upper right corner shows bubbles caused by insufficient pressure holding. Bubbles caused by high moisture content usually appear as large and irregularly distributed bubbles inside the product. Sometimes the bubbles are concentrated in the thick wall or deep inside the product. The edges of these bubbles may be fuzzy because the gas generated by the evaporation of water diffuses relatively freely in the melt, and the size and shape of the bubbles formed are irregular. Bubbles caused by the volatilization of additives: generally small and dispersed bubbles, which may be evenly distributed throughout the product or more near the surface of the product. Since the amount of gas generated by the volatilization of additives is relatively small and relatively uniform, the bubbles formed are small and relatively evenly distributed. Bubbles caused by poor exhaust: mostly concentrated in the last filling position of the mold, such as the edge of the product, near the gate, or the end away from the gate. The shape of the bubbles is usually round or oval, of different sizes, and sometimes appears in clusters. Because these positions are where gas is most likely to gather and the most difficult to discharge, the gas is compressed into bubbles in these parts. Bubbles caused by uneven mold temperature: often appear at the junction of faster and slower cooling of the product, and the bubbles are irregular in shape and inconsistent in size. In areas with rapid cooling, the gas in the plastic melt may be trapped due to rapid solidification, forming bubbles. For example, when injecting a product with a complex structure, if the mold temperature is not well controlled at the junction of thin walls and thick walls, bubbles may appear on the thick wall side. Bubbles caused by too fast an injection speed: Generally, they are randomly distributed on the surface or inside of the product. The bubbles are small and numerous and are dispersed. Because rapid injection causes a large amount of air to be drawn into the melt, these air forms tiny bubbles in the melt and disperses. Bubbles caused by insufficient holding pressure usually appear in the thick-walled area or inside the product, appearing as larger, isolated bubbles, sometimes accompanied by shrinkage marks. Etc. Step 002: Input the injection molded product bubble sample into the image recognition algorithm to obtain the predicted bubble type.

[0027] The image recognition algorithm may be a convolutional neural network, which includes a convolutional layer, an intermediate layer, a fully connected layer, and an output layer. The convolutional layer is used to extract image features from the surface image, thereby generating feature maps of multiple scales. Downsampling and pooling operations are performed through the intermediate layer, and the information in the feature map is further integrated through the fully connected layer. The Softmax activation function is used to output the probability of each category in the last layer of the fully connected layer. The category with the highest probability is the category to which the model predicts the image to belong, thereby obtaining the final predicted bubble type.

[0028] Step 003: Calculate the loss of the image recognition algorithm based on the predicted bubble type and the bubble type label.

[0029] Loss calculation is the process of measuring the difference between the model prediction result and the true label. The smaller the difference, the more accurate the model's prediction. In the implementation of the present invention, the loss is calculated by the mean square error loss function: in: is the value of the loss function. is the true sample label of the i-th sample. is the probability value of the bubble type predicted by the model for the i-th sample, and n is the number of samples. Step 004: Adjust the parameters of the image recognition algorithm according to the loss, and continue to input the injection molded product sample into the image recognition algorithm after the adjusted parameters for training, and continue to calculate the loss according to the new predicted bubble type and the bubble type label, until the loss is less than the preset loss threshold, and obtain a trained bubble recognition model.

[0030] Since bubbles at different positions in a sample image of bubbles of an injection molded product may correspond to different causes of bubble formation, in an embodiment of the present invention, the surface image is input into a bubble recognition model to obtain a bubble recognition result, which also specifically includes: inputting the surface image into the bubble recognition model to obtain at least one target bubble type; and determining the bubble recognition result according to the at least one target bubble type.

[0031] Step 130: When the bubble recognition result is that bubbles exist, the bubble image features in the surface image are obtained.

[0032] Among them, since the image recognition algorithm is a convolutional neural network, the convolutional neural network includes a convolutional layer, a fully connected layer, an intermediate layer and an output layer. The convolutional layer is used to extract features, and the intermediate layer is further downsampled and pooled. Therefore, in the embodiment of the present invention, when the bubble recognition result is that bubbles exist, the output of the intermediate layer of the convolutional neural network is used as the bubble image feature.

[0033] Step 140: Input the bubble image features and the current operating parameters into a bubble formation analysis model to obtain the bubble formation cause and bubble elimination strategy of the current injection molded product.

[0034] Among them, the bubble cause analysis model is obtained by inputting the bubble parameter samples into the multimodal fusion network for training; the bubble parameter samples include bubble image feature samples and corresponding operation parameter samples, as well as corresponding bubble cause labels and bubble elimination strategy labels; one bubble cause label corresponds to one bubble elimination strategy label. Among them, the specific training process of the bubble cause analysis model is generally consistent with the training process of the aforementioned image recognition algorithm. First, by obtaining the bubble parameter samples, the image feature samples and the corresponding operation parameter samples are preprocessed, and then spliced ​​together to obtain the spliced ​​feature sample data, and then input into the multimodal fusion network for fusion classification to obtain the predicted bubble cause and corresponding strategy. Then, the loss is calculated based on the predicted bubble cause and corresponding strategy and the bubble cause and corresponding strategy label, and the multimodal fusion network parameters are continuously adjusted according to the calculated loss, so as to obtain the final trained bubble cause analysis model.

[0035] Wherein, after the bubble formation analysis model is obtained through training, the bubble formation analysis model splices the bubble image features and the current operating parameters to obtain splicing feature data; the bubble formation analysis model fuses and classifies the splicing feature data to obtain the bubble formation cause and bubble elimination strategy of the current injection molding product. Wherein, the bubble elimination strategy includes at least one control parameter adjustment strategy, at least one equipment operation parameter adjustment strategy and at least one material adjustment strategy, the control parameter adjustment strategy includes exhaust control strategy, temperature control, injection speed control strategy, pressure holding control strategy, cooling time strategy, the equipment operation parameter adjustment strategy includes equipment screw adjustment strategy and equipment barrel adjustment strategy, and the material adjustment strategy includes water adjustment strategy and additive adjustment strategy. For example, when there are too many water bubbles, the corresponding water adjustment strategy; injection speed bubbles correspond to injection speed adjustment strategy; temperature bubbles correspond to temperature control adjustment strategy; insufficient pressure holding bubbles correspond to pressure holding control strategy, etc. Wherein, the bubble elimination strategy includes a preset strategy adjustment algorithm.

[0036] Step 150: Adjusting the processing of the injection molding equipment according to the bubble elimination strategy.

[0037] Among them, after obtaining the corresponding cause of bubble formation, the strategy adjustment algorithm preset in the bubble elimination strategy can be calculated according to the current equipment parameters or material parameters to obtain the final bubble elimination strategy. For example, when the bubble is caused by excessive water bubbles, the water adjustment algorithm in the corresponding water adjustment strategy and the current material ratio are calculated to adjust the water distribution ratio to obtain the final water distribution ratio, and the water distribution ratio parameters are sent to the injection molding equipment control platform so that the staff can adjust the water distribution ratio. When the bubble is caused by injection speed bubbles, the speed adjustment algorithm in the injection speed adjustment strategy and the current injection speed are calculated to obtain the final injection speed, and the injection speed parameters are sent to the injection molding equipment to control the injection speed.

[0038] The embodiment of the present invention obtains the current operating parameters of the injection molding equipment and the surface image of the current injection molded product taken at the outlet of the injection molding equipment, and inputs the surface image into the bubble recognition model to obtain the bubble recognition result; when the bubble recognition result is the presence of bubbles, the bubble image features in the surface image are obtained; the bubble image features and the current operating parameters are input into the bubble cause analysis model to obtain the bubble cause and bubble elimination strategy of the current injection molded product; the processing process of the injection molding equipment is adjusted according to the bubble elimination strategy. In this way, various types of surface bubbles can be accurately identified, so as to trace the cause of bubbles in the processing process, so as to control the processing process of the injection molded product and reduce the probability of defective injection molded products. In this way, the embodiment of the present invention greatly improves the intelligence of the processing control of the injection molding equipment, so that the surface defects of the injection molded product can be identified more quickly and accurately, and can be adjusted quickly and accurately.

[0039] Figure 3 The schematic diagram of the structure of the processing control device of the injection molding product provided by the embodiment of the present invention is shown. Figure 3 As shown, the device 300 includes: a first acquisition module 310, which is used to acquire the current operating parameters of the injection molding equipment and the surface image of the current injection molding product photographed at the outlet of the injection molding equipment; wherein an image shooting cover is provided at the outlet of the injection molding equipment, and a camera device and a plurality of light sources at different angles and color temperatures are provided in the image shooting cover; the camera device is used to acquire the surface image of the current injection molding product; The recognition module 320 is used to input the surface image into a bubble recognition model to obtain a bubble recognition result; the bubble recognition result includes the presence or absence of bubbles; the bubble recognition model is obtained by training injection molded product bubble samples according to an image recognition algorithm, and the injection molded product bubble samples include multiple types of injection molded product bubble sample images and corresponding bubble type labels; A second acquisition module 330 is used to acquire bubble image features in the surface image when the bubble recognition result is that bubbles exist; The analysis module 340 is used to input the bubble image features and the current operating parameters into the bubble cause analysis model to obtain the bubble cause and bubble elimination strategy of the current injection molded product; the bubble cause analysis model is obtained by inputting a multimodal fusion network training according to bubble parameter samples; the bubble parameter samples include bubble image feature samples and corresponding operating parameter samples, as well as corresponding bubble cause labels and bubble elimination strategy labels; The adjustment module 350 is used to adjust the processing process of the injection molding equipment according to the bubble elimination strategy.

[0040] In an optional manner, the obtaining of the current operating parameters of the injection molding equipment and the surface image of the current injection molded product photographed at the outlet of the injection molding equipment includes: Acquire the real-time control parameters of the control system of the injection molding equipment and the real-time operation parameters of the equipment when processing the current injection molding product as the current operation parameters; the real-time control parameters include injection speed, holding pressure, cooling time and mold temperature; the real-time operation parameters of the equipment include the temperature, pressure, motor speed and hydraulic system pressure of each equipment component; When the injection molding of the current injection-molded product is completed, the surface images of the current injection-molded product at multiple angles are acquired by the photographing device.

[0041] In an optional manner, the device further includes: A sample acquisition module is used to acquire injection molded product bubble samples; the injection molded product bubble samples include multiple types of injection molded product bubble sample images and corresponding bubble type labels; the bubble type labels include water bubbles, additive bubbles, poor exhaust bubbles, temperature uneven bubbles, injection speed bubbles, insufficient pressure holding bubbles, cooling time bubbles, equipment screw wear bubbles and equipment barrel temperature bubbles; A training prediction module is used to input the injection molded product bubble sample into the image recognition algorithm to obtain a predicted bubble type; a loss calculation module, used to calculate the loss of the image recognition algorithm according to the predicted bubble type and the bubble type label; A model adjustment module, used for adjusting the parameters of the image recognition algorithm according to the loss, continuing to input the injection molded product sample into the image recognition algorithm after the parameter adjustment for training, and continuing to calculate the loss according to the new predicted bubble type and the bubble type label, until the loss is less than a preset loss threshold, thereby obtaining a trained bubble recognition model; The step of inputting the surface image into a bubble recognition model to obtain a bubble recognition result includes: Inputting the surface image into a bubble recognition model to obtain a target bubble type; The bubble recognition result is determined according to the at least one target bubble type.

[0042] In an optional manner, the image recognition algorithm is a convolutional neural network, and the convolutional neural network includes a convolutional layer, a fully connected layer, and an intermediate layer; when the bubble recognition result is that bubbles exist, obtaining bubble image features in the surface image includes: When the bubble recognition result is that bubbles exist, the output of the middle layer of the convolutional neural network is used as the bubble image feature.

[0043] In an optional manner, the bubble image features and the current operating parameters are input into a bubble formation analysis model to obtain the bubble formation cause and bubble elimination strategy of the current injection molded product, including: Splicing the bubble image features and the current operating parameters to obtain splicing feature data; The splicing feature data are fused and classified to obtain the bubble cause and bubble elimination strategy of the current injection molding product.

[0044] In an optional manner, the elimination strategy includes at least one control parameter adjustment strategy, at least one equipment operation parameter adjustment strategy and at least one material adjustment strategy. The control parameter adjustment strategy includes exhaust control strategy, temperature control, injection speed control strategy, pressure holding control strategy and cooling time strategy. The equipment operation parameter adjustment strategy includes equipment screw adjustment strategy and equipment barrel adjustment strategy. The material adjustment strategy includes moisture adjustment strategy and additive adjustment strategy.

[0045] In an optional manner, the obtaining of the current operating parameters of the injection molding equipment and the surface image of the current injection molded product photographed at the outlet of the injection molding equipment includes: Decrypting the current operating parameters according to a preset decryption method to obtain the decrypted current operating parameters; The decrypted current operating parameters are normalized to obtain normalized current operating parameters.

[0046] The embodiment of the present invention obtains the current operating parameters of the injection molding equipment and the surface image of the current injection molded product taken at the outlet of the injection molding equipment, and inputs the surface image into a bubble recognition model to obtain a bubble recognition result; when the bubble recognition result is that bubbles exist, the bubble image features in the surface image are obtained; the bubble image features and the current operating parameters are input into a bubble cause analysis model to obtain the bubble cause and bubble elimination strategy of the current injection molded product; the processing process of the injection molding equipment is adjusted according to the bubble elimination strategy. In this way, various types of surface bubbles can be accurately identified, thereby tracing the causes of bubbles in the processing process, so that the processing process of the injection molded product can be controlled and the probability of defective injection molded products can be reduced.

[0047] Figure 4 The schematic diagram of the structure of the processing control device for injection molding products provided by the embodiment of the present invention is shown. The specific embodiment of the present invention does not limit the specific implementation of the processing control device for injection molding products.

[0048] like Figure 4 As shown, the processing control device of the injection molding product may include: a processor (processor) 402, a communication interface (Communications Interface) 404, a memory (memory) 406, and a communication bus 408.

[0049] The processor 402, the communication interface 404, and the memory 406 communicate with each other via the communication bus 408. The communication interface 404 is used to communicate with other devices such as a client or other server network elements. The processor 402 is used to execute the program 410, which can specifically execute the relevant steps in the above-mentioned embodiment of the processing control method for injection molding products.

[0050] Specifically, the program 410 may include program code including computer executable instructions.

[0051] The processor 402 may be a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiment of the present invention. The one or more processors included in the processing control device for injection molding products may be processors of the same type, such as one or more CPUs; or may be processors of different types, such as one or more CPUs and one or more ASICs.

[0052] The memory 406 is used to store the program 410. The memory 406 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0053] Program 410 can be specifically called by processor 402 to enable the processing control device of the injection molding product to perform the following operations: Acquire the current operating parameters of the injection molding equipment and the surface image of the current injection molding product captured at the injection molding equipment outlet; wherein an image capturing hood is provided at the injection molding equipment outlet, and a camera device and a plurality of light sources at different angles and color temperatures are provided in the image capturing hood; the camera device is used to acquire the surface image of the current injection molding product; The surface image is input into a bubble recognition model to obtain a bubble recognition result; the bubble recognition result includes the presence or absence of bubbles; the bubble recognition model is obtained by training injection molded product bubble samples according to an image recognition algorithm, and the injection molded product bubble samples include multiple types of injection molded product bubble sample images and corresponding bubble type labels; When the bubble recognition result is that bubbles exist, acquiring bubble image features in the surface image; The bubble image features and the current operating parameters are input into the bubble cause analysis model to obtain the bubble cause and bubble elimination strategy of the current injection molded product; the bubble cause analysis model is obtained by inputting a multimodal fusion network training based on bubble parameter samples; the bubble parameter samples include bubble image feature samples and corresponding operating parameter samples, as well as corresponding bubble cause labels and bubble elimination strategy labels; The processing of the injection molding equipment is adjusted according to the bubble elimination strategy.

[0054] In an optional manner, the obtaining of the current operating parameters of the injection molding equipment and the surface image of the current injection molded product photographed at the outlet of the injection molding equipment includes: Acquire the real-time control parameters of the control system of the injection molding equipment and the real-time operation parameters of the equipment when processing the current injection molding product as the current operation parameters; the real-time control parameters include injection speed, holding pressure, cooling time and mold temperature; the real-time operation parameters of the equipment include the temperature, pressure, motor speed and hydraulic system pressure of each equipment component; When the injection molding of the current injection-molded product is completed, the surface images of the current injection-molded product at multiple angles are acquired by the photographing device.

[0055] In an optional manner, before inputting the surface image into the bubble recognition model to obtain the bubble recognition result, the method includes: Acquire injection molded product bubble samples; the injection molded product bubble samples include multiple types of injection molded product bubble sample images and corresponding bubble type labels; the bubble type labels include water bubbles, additive bubbles, poor exhaust bubbles, temperature uneven bubbles, injection speed bubbles, insufficient pressure holding bubbles, cooling time bubbles, equipment screw wear bubbles and equipment barrel temperature bubbles; Inputting the injection molded product bubble sample into the image recognition algorithm to obtain a predicted bubble type; Calculating a loss of the image recognition algorithm based on the predicted bubble type and the bubble type label; Adjusting the parameters of the image recognition algorithm according to the loss, and continuing to input the injection molded product sample into the image recognition algorithm after the adjusted parameters for training, and continuing to calculate the loss according to the new predicted bubble type and the bubble type label, until the loss is less than a preset loss threshold, thereby obtaining a trained bubble recognition model; The step of inputting the surface image into a bubble recognition model to obtain a bubble recognition result includes: Inputting the surface image into a bubble recognition model to obtain a target bubble type; The bubble recognition result is determined according to the at least one target bubble type.

[0056] In an optional manner, the image recognition algorithm is a convolutional neural network, and the convolutional neural network includes a convolutional layer, a fully connected layer, and an intermediate layer; when the bubble recognition result is that bubbles exist, obtaining bubble image features in the surface image includes: When the bubble recognition result is that bubbles exist, the output of the middle layer of the convolutional neural network is used as the bubble image feature.

[0057] In an optional manner, the bubble image features and the current operating parameters are input into a bubble formation analysis model to obtain the bubble formation cause and bubble elimination strategy of the current injection molded product, including: Splicing the bubble image features and the current operating parameters to obtain splicing feature data; The splicing feature data are fused and classified to obtain the bubble cause and bubble elimination strategy of the current injection molding product.

[0058] In an optional manner, the elimination strategy includes at least one control parameter adjustment strategy, at least one equipment operation parameter adjustment strategy and at least one material adjustment strategy. The control parameter adjustment strategy includes exhaust control strategy, temperature control, injection speed control strategy, pressure holding control strategy and cooling time strategy. The equipment operation parameter adjustment strategy includes equipment screw adjustment strategy and equipment barrel adjustment strategy. The material adjustment strategy includes moisture adjustment strategy and additive adjustment strategy.

[0059] In an optional manner, the obtaining of the current operating parameters of the injection molding equipment and the surface image of the current injection molded product photographed at the outlet of the injection molding equipment includes: Decrypting the current operating parameters according to a preset decryption method to obtain the decrypted current operating parameters; The decrypted current operating parameters are normalized to obtain normalized current operating parameters.

[0060] The embodiment of the present invention obtains the current operating parameters of the injection molding equipment and the surface image of the current injection molded product taken at the outlet of the injection molding equipment, and inputs the surface image into a bubble recognition model to obtain a bubble recognition result; when the bubble recognition result is that bubbles exist, the bubble image features in the surface image are obtained; the bubble image features and the current operating parameters are input into a bubble cause analysis model to obtain the bubble cause and bubble elimination strategy of the current injection molded product; the processing process of the injection molding equipment is adjusted according to the bubble elimination strategy. In this way, various types of surface bubbles can be accurately identified, thereby tracing the causes of bubbles in the processing process, so that the processing process of the injection molded product can be controlled and the probability of defective injection molded products can be reduced.

[0061] An embodiment of the present invention provides a computer-readable storage medium, which stores at least one executable instruction. When the executable instruction is executed on a processing control device for an injection molding product, the processing control device for the injection molding product executes the processing control method for the injection molding product in any of the above method embodiments.

[0062] The executable instructions can be specifically used to enable the processing control equipment of the injection molding product to perform the following operations: Acquire the current operating parameters of the injection molding equipment and the surface image of the current injection molding product captured at the injection molding equipment outlet; wherein an image capturing hood is provided at the injection molding equipment outlet, and a camera device and a plurality of light sources at different angles and color temperatures are provided in the image capturing hood; the camera device is used to acquire the surface image of the current injection molding product; The surface image is input into a bubble recognition model to obtain a bubble recognition result; the bubble recognition result includes the presence or absence of bubbles; the bubble recognition model is obtained by training injection molded product bubble samples according to an image recognition algorithm, and the injection molded product bubble samples include multiple types of injection molded product bubble sample images and corresponding bubble type labels; When the bubble recognition result is that bubbles exist, acquiring bubble image features in the surface image; The bubble image features and the current operating parameters are input into the bubble cause analysis model to obtain the bubble cause and bubble elimination strategy of the current injection molded product; the bubble cause analysis model is obtained by inputting a multimodal fusion network training based on bubble parameter samples; the bubble parameter samples include bubble image feature samples and corresponding operating parameter samples, as well as corresponding bubble cause labels and bubble elimination strategy labels; The processing of the injection molding equipment is adjusted according to the bubble elimination strategy.

[0063] In an optional manner, the obtaining of the current operating parameters of the injection molding equipment and the surface image of the current injection molded product photographed at the outlet of the injection molding equipment includes: Acquire the real-time control parameters of the control system of the injection molding equipment and the real-time operation parameters of the equipment when processing the current injection molding product as the current operation parameters; the real-time control parameters include injection speed, holding pressure, cooling time and mold temperature; the real-time operation parameters of the equipment include the temperature, pressure, motor speed and hydraulic system pressure of each equipment component; When the injection molding of the current injection-molded product is completed, the surface images of the current injection-molded product at multiple angles are acquired by the photographing device.

[0064] In an optional manner, before inputting the surface image into the bubble recognition model to obtain the bubble recognition result, the method includes: Acquire injection molded product bubble samples; the injection molded product bubble samples include multiple types of injection molded product bubble sample images and corresponding bubble type labels; the bubble type labels include water bubbles, additive bubbles, poor exhaust bubbles, temperature uneven bubbles, injection speed bubbles, insufficient pressure holding bubbles, cooling time bubbles, equipment screw wear bubbles and equipment barrel temperature bubbles; Inputting the injection molded product bubble sample into the image recognition algorithm to obtain a predicted bubble type; Calculating a loss of the image recognition algorithm based on the predicted bubble type and the bubble type label; Adjusting the parameters of the image recognition algorithm according to the loss, and continuing to input the injection molded product sample into the image recognition algorithm after the adjusted parameters for training, and continuing to calculate the loss according to the new predicted bubble type and the bubble type label, until the loss is less than a preset loss threshold, thereby obtaining a trained bubble recognition model; The step of inputting the surface image into a bubble recognition model to obtain a bubble recognition result includes: Inputting the surface image into a bubble recognition model to obtain a target bubble type; The bubble recognition result is determined according to the at least one target bubble type.

[0065] In an optional manner, the image recognition algorithm is a convolutional neural network, and the convolutional neural network includes a convolutional layer, a fully connected layer, and an intermediate layer; when the bubble recognition result is that bubbles exist, obtaining bubble image features in the surface image includes: When the bubble recognition result is that bubbles exist, the output of the middle layer of the convolutional neural network is used as the bubble image feature.

[0066] In an optional manner, the bubble image features and the current operating parameters are input into a bubble formation analysis model to obtain the bubble formation cause and bubble elimination strategy of the current injection molded product, including: Splicing the bubble image features and the current operating parameters to obtain splicing feature data; The splicing feature data are fused and classified to obtain the bubble cause and bubble elimination strategy of the current injection molding product.

[0067] In an optional manner, the elimination strategy includes at least one control parameter adjustment strategy, at least one equipment operation parameter adjustment strategy and at least one material adjustment strategy. The control parameter adjustment strategy includes exhaust control strategy, temperature control, injection speed control strategy, pressure holding control strategy and cooling time strategy. The equipment operation parameter adjustment strategy includes equipment screw adjustment strategy and equipment barrel adjustment strategy. The material adjustment strategy includes moisture adjustment strategy and additive adjustment strategy.

[0068] In an optional manner, the obtaining of the current operating parameters of the injection molding equipment and the surface image of the current injection molded product photographed at the outlet of the injection molding equipment includes: Decrypting the current operating parameters according to a preset decryption method to obtain the decrypted current operating parameters; The decrypted current operating parameters are normalized to obtain normalized current operating parameters.

[0069] The embodiment of the present invention obtains the current operating parameters of the injection molding equipment and the surface image of the current injection molded product taken at the outlet of the injection molding equipment, and inputs the surface image into a bubble recognition model to obtain a bubble recognition result; when the bubble recognition result is that bubbles exist, the bubble image features in the surface image are obtained; the bubble image features and the current operating parameters are input into a bubble cause analysis model to obtain the bubble cause and bubble elimination strategy of the current injection molded product; the processing process of the injection molding equipment is adjusted according to the bubble elimination strategy. In this way, various types of surface bubbles can be accurately identified, thereby tracing the causes of bubbles in the processing process, so that the processing process of the injection molded product can be controlled and the probability of defective injection molded products can be reduced.

[0070] The algorithm or display provided herein is not inherently related to any particular computer, virtual system or other equipment. Various general purpose systems can also be used together with the teachings based on this. According to the above description, it is obvious to construct the structure required for this type of system. In addition, the embodiment of the present invention is not directed to any specific programming language yet. It should be understood that various programming languages ​​can be utilized to realize the content of the present invention described herein, and the description made to specific languages ​​above is for disclosing the best mode of the present invention.

[0071] In the description provided herein, a large number of specific details are described. However, it is understood that embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures and techniques are not shown in detail so as not to obscure the understanding of this description.

[0072] Similarly, it should be understood that in order to streamline the present invention and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of the present invention, various features of the embodiments of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof. However, this method of disclosure should not be interpreted as reflecting an intention that the claimed invention requires more features than those expressly recited in each claim.

[0073] Those skilled in the art will appreciate that the modules in the devices in the embodiments may be adaptively changed and set in one or more devices different from the embodiments. The modules or units or components in the embodiments may be combined into one module or unit or component, and may be divided into multiple submodules or subunits or subcomponents. All features disclosed in this specification (including the accompanying claims, abstracts and drawings) and all processes or units of any method or device disclosed in this manner may be combined in any combination, except that at least some of such features and / or processes or units are mutually exclusive. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstracts and drawings) may be replaced by an alternative feature that provides the same, equivalent or similar purpose.

[0074] It should be noted that the above embodiments illustrate the present invention rather than limit it, and that those skilled in the art may devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference symbol between brackets shall not be construed as a limitation on the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "one" or "an" preceding an element does not exclude the presence of a plurality of such elements. The present invention may be implemented by means of hardware comprising a number of different elements and by means of a suitably programmed computer. In a unit claim enumerating a number of devices, several of these devices may be embodied by the same hardware item. The use of the words first, second, and third, etc. does not indicate any order. These words may be interpreted as names. The steps in the above embodiments, unless otherwise specified, should not be understood as limitations on the order of execution.

Claims

1. A processing control method for injection molded products, characterized in that: The method comprises: Acquire the current operating parameters of the injection molding equipment and the surface image of the current injection molding product captured at the injection molding equipment outlet; wherein an image capturing hood is provided at the injection molding equipment outlet, and a camera device and a plurality of light sources at different angles and color temperatures are provided in the image capturing hood; the camera device is used to acquire the surface image of the current injection molding product; The surface image is input into a bubble recognition model to obtain a bubble recognition result; the bubble recognition result includes the presence or absence of bubbles; the bubble recognition model is obtained by training injection molded product bubble samples according to an image recognition algorithm, and the injection molded product bubble samples include multiple types of injection molded product bubble sample images and corresponding bubble type labels; When the bubble recognition result is that bubbles exist, acquiring bubble image features in the surface image; The bubble image features and the current operating parameters are input into the bubble cause analysis model to obtain the bubble cause and bubble elimination strategy of the current injection molded product; the bubble cause analysis model is obtained by inputting a multimodal fusion network training based on bubble parameter samples; the bubble parameter samples include bubble image feature samples and corresponding operating parameter samples, as well as corresponding bubble cause labels and bubble elimination strategy labels; The processing of the injection molding equipment is adjusted according to the bubble elimination strategy.

2. The method according to claim 1, characterized in that The method of obtaining the current operating parameters of the injection molding equipment and the surface image of the current injection molded product photographed at the outlet of the injection molding equipment includes: Acquire the real-time control parameters of the control system of the injection molding equipment and the real-time operation parameters of the equipment when processing the current injection molding product as the current operation parameters; the real-time control parameters include injection speed, holding pressure, cooling time and mold temperature; the real-time operation parameters of the equipment include the temperature, pressure, motor speed and hydraulic system pressure of each equipment component; When the injection molding of the current injection-molded product is completed, the surface images of the current injection-molded product at multiple angles are acquired by the photographing device.

3. The method according to claim 2, characterized in that Before inputting the surface image into the bubble recognition model to obtain the bubble recognition result, the method includes: Acquire injection molded product bubble samples; the injection molded product bubble samples include multiple types of injection molded product bubble sample images and corresponding bubble type labels; the bubble type labels include water bubbles, additive bubbles, poor exhaust bubbles, temperature uneven bubbles, injection speed bubbles, insufficient pressure holding bubbles, cooling time bubbles, equipment screw wear bubbles and equipment barrel temperature bubbles; Inputting the injection molded product bubble sample into the image recognition algorithm to obtain a predicted bubble type; Calculating a loss of the image recognition algorithm based on the predicted bubble type and the bubble type label; Adjusting the parameters of the image recognition algorithm according to the loss, and continuing to input the injection molded product sample into the image recognition algorithm after the adjusted parameters for training, and continuing to calculate the loss according to the new predicted bubble type and the bubble type label, until the loss is less than a preset loss threshold, thereby obtaining a trained bubble recognition model; The step of inputting the surface image into a bubble recognition model to obtain a bubble recognition result includes: Inputting the surface image into a bubble recognition model to obtain at least one target bubble type; The bubble recognition result is determined according to the at least one target bubble type.

4. The method according to any one of claims 1 to 3, characterized in that: The image recognition algorithm is a convolutional neural network, and the convolutional neural network includes a convolutional layer, a fully connected layer, and an intermediate layer; when the bubble recognition result is that bubbles exist, the bubble image features in the surface image are obtained, including: When the bubble recognition result is that bubbles exist, the output of the middle layer of the convolutional neural network is used as the bubble image feature.

5. The method according to any one of claims 1 to 3, characterized in that: The step of inputting the bubble image features and the current operating parameters into a bubble formation analysis model to obtain the bubble formation cause and bubble elimination strategy of the current injection molded product includes: The bubble formation analysis model splices the bubble image features and the current operating parameters to obtain splicing feature data; The bubble formation analysis model fuses and classifies the splicing feature data to obtain the bubble formation cause and bubble elimination strategy of the current injection molding product.

6. The method according to any one of claims 1 to 3, characterized in that: The elimination strategy includes at least one control parameter adjustment strategy, at least one equipment operation parameter adjustment strategy and at least one material adjustment strategy. The control parameter adjustment strategy includes exhaust control strategy, temperature control, injection speed control strategy, pressure holding control strategy and cooling time strategy. The equipment operation parameter adjustment strategy includes equipment screw adjustment strategy and equipment barrel adjustment strategy. The material adjustment strategy includes moisture adjustment strategy and additive adjustment strategy.

7. The method according to any one of claims 1 to 3, characterized in that: The method of obtaining the current operating parameters of the injection molding equipment and the surface image of the current injection molded product photographed at the outlet of the injection molding equipment includes: Decrypting the current operating parameters according to a preset decryption method to obtain the decrypted current operating parameters; The decrypted current operating parameters are normalized to obtain normalized current operating parameters.

8. A processing control device for injection molded products, characterized in that: The device comprises: A first acquisition module is used to acquire the current operating parameters of the injection molding equipment and the surface image of the current injection molding product captured at the outlet of the injection molding equipment; wherein an image capturing hood is provided at the outlet of the injection molding equipment, and a camera device and a plurality of light sources at different angles and color temperatures are provided in the image capturing hood; the camera device is used to acquire the surface image of the current injection molding product; A recognition module, used for inputting the surface image into a bubble recognition model to obtain a bubble recognition result; the bubble recognition result includes the presence or absence of bubbles; the bubble recognition model is obtained by training injection molded product bubble samples according to an image recognition algorithm, and the injection molded product bubble samples include multiple types of injection molded product bubble sample images and corresponding bubble type labels; A second acquisition module, configured to acquire bubble image features in the surface image when the bubble recognition result indicates that bubbles exist; An analysis module is used to input the bubble image features and the current operating parameters into a bubble cause analysis model to obtain the bubble cause and bubble elimination strategy of the current injection molded product; the bubble cause analysis model is obtained by inputting a multimodal fusion network training based on bubble parameter samples; the bubble parameter samples include bubble image feature samples and corresponding operating parameter samples, as well as corresponding bubble cause labels and bubble elimination strategy labels; An adjustment module is used to adjust the processing process of the injection molding equipment according to the bubble elimination strategy.

9. A processing control device for injection molded products, characterized in that: include: A processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to execute the operation of the processing control method of the injection molding product as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The storage medium stores at least one executable instruction. When the executable instruction is executed on the processing control device of the injection molding product, the processing control device of the injection molding product executes the operation of the processing control method of the injection molding product according to any one of claims 1 to 7.

Images